Reviving battery leak damaged IIfx, sad mac crash chime

[AppleTree]

Hi

I bought a non-working Mac IIfx from eBay.  There was a battery leak damage to the motherboard.  Some components near the start up circuit were missing and many traces and vias damaged.  The SRAM at UD16 had a couple of pins corroded away as well.  Not surprisingly, initially it didn't boot at all.

I cleaned the motherboard in alcohol (I soaked the board in it) and replaced missing components.  Replaced the leaky caps.  And I replaced the SRAM with a NOS from eBay (The board had M5M5256BFP but I replaced it with equivalent HM62256LFP, which is actually the chip on the IIfx schematic).  I got it to boot with normal chime but after the screen comes on, it gives a crash chime but no error code.  It just hangs on the grey screen.  (Please see the video below - in the video I put the RAM chips in Bank B, but it makes no difference whether the RAM chips are in Bank A or Bank B.  Behaviour is exactly the same)

I replaced the ROM with BMOW Rominator and it also starts with normal chime, screen comes on, followed by a crash chime (same as above) but interestingly this one gives a Sad Mac error code 0000000F 00000001.

I don't know how far it gets in self test but it seems to get past the memory test, as when I insert 16MB as opposed to 4MB, the crash chime takes longer to appear (using the original ROM)

I've fixed all the damaged vias and traces that I can see around the batteries.

I used the IIfx schematic to check the connections near the battery as the following:

- All pins going out from top RAM slot in Bank A (closest slot to the batteries)

- All pins going out from SRAM (UD16) - The one I replaced

- All connections between Z8530 (UC17) and Serial PIC (UC16)

- All connections between RTC (UF18) and 65C23 (UE12)

I am stumped as to what to do from here.  I presume Rominator would give the same error code as the Original ROM, though I am not sure why the original ROM does not show the error code.  What is 0000000F 00000001?  The web page I found is a little confusing to understand.  It says Mac II error with 000F ending on the first line is reserved for compatibility.  What does that mean?  Closest error code I can find on the page is actually (000F 0001) Bus Error.

Any idea what I should test next?  I have a working IIfx so I am going to swap the CPU tomorrow to see if that makes any difference, though I doubt if the CPU was faulty, it would even get that far.

Could the damaged Z8530 (serial comm controller above the battery) cause an error?  That is another chip that had bit suspicious looking pins from battery damage, though after cleaning with alcohol, it looks ok).

Thanks in advance.


View attachment iifx-sad-chime.mp4

 

[techknight]

Thats a stuck NMI button, or floating NMI. 

Look at the NMI button traces. Check for shorts, or opens. Especially the pull-up resistor. R11 being missing may have something to do with this. your missing a few components in that area. 

 

[AppleTree]

Thats a stuck NMI button, or floating NMI. 

Look at the NMI button traces. Check for shorts, or opens. Especially the pull-up resistor. R11 being missing may have something to do with this. your missing a few components in that area. 

Thank you @techknight for your reply.

The first photo was just to show what the board looked like when I got it.  As part of repair, I cleaned up the area and added all of the missing components including R11.  I don't believe NMI button is stuck because if I press it before the screen comes on, it goes to death chime straight away.  So I know the button functions and it registers when I press it.  However I will check the NMI traces again as you suggested.

I checked all traces from four Bank A RAM slots to RP6,7,8 and UL11,13,16,17 (74F573) yesterday using the schematic and they are all good so it doesn't look to be trace issues with SIMM slots.  (BTW I found a mistake in the Bomarc IIfx schematic, SIMM slot data out pin should be 47, not 44.  I found a couple more as well, which I can't recall right now)

So it looks like POST passes and it goes to RAM check and I thought it maybe failing there but with all SIMM slot traces looking good, that is not the case.

From various places I read that RAM check is the last check Mac does before showing the happy Mac screen.  If RAM SIMM traces are all good, why would it fail?  Could something between RAM SIMM and the memory controller be failing?
 

I am also going to hook up another Mac and see if I can use the diagnostic mode to get some more info.  Never used a diagnostic mode before but hopefully I will figure out how to use it.

I added two more photos.  One showing after cleaning up the damaged area and another showing after missing components added in (R11, D2, D3, R18)

Thanks.

 

[techknight]

I dont know. All I know is the 0F 01 is usually the NMI. 

 

[AppleTree]

I dont know. All I know is the 0F 01 is usually the NMI. 

Thanks @techknight.  Yes the online resources tell me 0000000F is software related, which confused me.  But maybe NMI is being triggered somewhere, causing this error code to appear.  I will check the NMI area carefully (which I haven't done yet apart from adding R11).

At this stage, I am assuming POST is successful (the fact initial boot chime sounds followed by the screen coming on and going to memory check) so it's either memory or NMI as you suggested.

Thanks.

 

[AppleTree]

Yeah, I've fixed it.  It turned out to be the RTC chip.

I checked the NMI and it was all fine so that was not the problem.  Then I searched the forum and found someone who had the exactly same issue as me last year, albeit with Q700.  Battery leak damage, startup chime is fine and then the computer hangs with 0000000F 00000001 error.  @AlexTheCat123 re-soldered the RTC chip and got Q700 to boot.

I actually had checked RTC chip connection a few days ago against the schematic so I was a bit skeptical that this would work but I was running out of options.  I de-soldered the chip and found that two pads were badly damaged.  I cleaned the board under the chip with alcohol and re-soldered the chip and WOW it fixed the issue.  IIfx is alive again.

As I mentioned, I had already checked the RTC chip connection so I am not sure what changed.  Maybe the two damaged pads weren't making good contact before or maybe there was cap goo under the chip that I could not clean with the chip soldered on.

So it looks like IIfx does POST and then RAM check and finally accesses the RTC/PRAM chip, which I presume is for getting the stored configuration setting (maybe get the startup disk config)?  When it fails, it hangs with 0F 01 error.

Thank you @techknight for replying to this post.  No one was answering to my post so I was getting a bit discouraged there so it was great to have your replies to keep me going.  I was getting close to just giving up.

Thank you @AlexTheCat123 for leaving your experience with 0F 01 error.  That was the last missing piece of the puzzle that I needed to find after replacing a SRAM chip and fixing about 10 traces and vias.


View attachment Mac_IIfx_-_fixed.mp4

 

[techknight]

I didnt even think about the RTC chip because ive seen machines run without it. I guess the IIfx is not one of them.

 

[ravuya]

Nice save. I've got a IIfx with double-Maxell damage that I need to get started on as well, so this gives me hope that mine could possibly be saved as well.